vidore/syntheticDocQA_artificial_intelligence_test_beir

该数据集是BEIR版本的vidore/syntheticDocQA_artificial_intelligence_test数据集，包含三个部分：corpus（文档信息），qrels（查询与文档的相关性评分），和queries（查询信息）。corpus部分包含文档的唯一标识符和图像，qrels部分包含查询的唯一标识符、文档的唯一标识符和评分，queries部分包含查询的唯一标识符和查询内容。数据集分为测试集，每个部分的大小和示例数量都有详细说明。

This dataset is the BEIR version of the vidore/syntheticDocQA_artificial_intelligence_test dataset, consisting of three parts: corpus (document information), qrels (relevance scores between queries and documents), and queries (query information). The corpus part includes a unique identifier and an image for each document, the qrels part includes a unique identifier for the query, a unique identifier for the document, and a score, and the queries part includes a unique identifier and the content of the query. The dataset is split into a test set, with detailed descriptions of the size and number of examples for each part.